Prediction of rotor-spun yarn quality using hybrid artificial neural network-fuzzy expert system model

Ghanmi, Hanen ; Ghith, Adel ; Benameur, Tarek


This study aims at developing a new approach to predict and determine the quality of rotor-spun yarn in terms of fibre characteristics as well as critical yarn properties. Hybrid modeling by combining two or more techniques has been demonstrated to give better performance than that of several single approaches over many research areas. Hence, in this study a hybrid model by combining two soft computing approaches, namely artificial neural network (ANN) and fuzzy expert system, has been developed. The ANN is used to predict three yarn characteristics, namely tenacity, breaking elongation and CVm. Then these three outputs are used to predict the new quality index by means of the fuzzy expert system. The accuracy of predicted model has been estimated using statistical performance criteria, such as correlation coefficient (R), root mean square error (RMSE), mean absolute error (MAE) and mean relative per cent error (MRPE). The results show the ability of model to predict the rotor-spun yarn quality and according to the analytical findings, the hybrid model gives accurate result.



Artificial neural network;Fuzzy expert system;Global yarn quality;Hybrid model;Rotor-spun yarn

Full Text: PDF (downloaded 699 times)


  • There are currently no refbacks.
This abstract viewed 566 times